ROOT-CAUSE ANALYSIS
AgentDebug
Collaborative, empirically grounded reasoning for root-cause analysis in microservice systems using operational evidence.
SOFTWARE ENGINEERING RESEARCHER
Mahsa
Panahandeh.PhD
Understanding complex systems and behaviours, making systems more reliable.
I am a postdoctoral fellow at the University of Ottawa, working with Lionel C. Briand at Nanda Lab. My research brings together software observability, AIOps, and agentic software engineering to diagnose failures and improve the reliability of complex software systems.
I completed my PhD in February 2025 at the University of Alberta in the Software Engineering and Intelligent Systems program, focusing on AIOps and performance engineering. My doctoral thesis explored anomaly detection and root cause localization in microservices using distributed traces and profiling metrics.
I am drawn to the challenge of bringing multiple sources of evidence together to understand systems, reason about their behavior, and improve their performance and reliability. My research connects operational evidence with AI to support more effective failure diagnosis, software debugging, and automated software engineering.
Current Research
AI-ASSISTED REPAIR
Agentic Test Repair
Specialized agents for repairing AI-generated test cases in CI/CD environments.
EMPIRICAL SOFTWARE ENGINEERING
Observability of AI-Generated Software
Studying whether AI-generated code follows human observability practices.
Research Interests
Observability & AIOps
Using distributed traces, metrics, and logs to detect anomalies and understand the behavior of microservice systems.
Performance Debugging
Identifying root causes of performance degradation through contextual analysis and interpretable diagnostic evidence.
AI for software engineering
Exploring how large language models can support fault localization, software repair, and reasoning about failures.
Selected Publications
arXiv preprint2026
Similar Pattern Annotation via Retrieval Knowledge for LLM-Based Test Code Fault Localization
SPARK · *Equal contribution.
FSE Companion2026
Rethinking Performance Debugging: From Optimization to Collaborative Reasoning
Vision paper · pp. 1312–1316
FSE · Industry Track2026
TraceSynth: Generating Production-Quality Kernel Traces with Constraint-Guided Diffusion Models
Industry paper · pp. 496–506
ICPE2026
B-Perf: Black-box Performance Antipattern Detection Using System-level Execution Tracing
Research paper · pp. 96–107
ICPE Companion2026
Representation-Aware Root Cause Analysis with Large Language Models
Position paper · pp. 62–68
IEEE TSE2026
CARE: Context Aware Root Cause Identification Using Distributed Traces and Profiling Metrics
Combining observability data with contextual analysis to localize performance root causes in microservices.
JSS2024
ServiceAnomaly: An anomaly detection approach in microservices using distributed traces and profiling metrics
Detecting abnormal microservice behavior using request execution paths and profiling data.
ICPE Companion2024
Context-aware Root Cause Localization in Distributed Traces Using Social Network Analysis
Work in progress · pp. 1–6
ICPE Companion2024
Efficient Unsupervised Latency Culprit Ranking in Distributed Traces with GNN and Critical Path Analysis
Data challenge paper · pp. 62–66
SoSyM2021
MUPPIT: A Method for Using Proper Patterns in Model Transformations
Software and Systems Modeling · 20, pp. 1491–1523
ASE Journal2018
Black-box Tree Test Case Generation Through Diversity
Automated Software Engineering · 25(3), pp. 531–568
Background
Postdoctoral Fellow
University of Ottawa · Nanda Lab (May 2025–present)
Leading research on AI-assisted fault localization, automated test repair, and performance diagnosis, including collaboration with a large industrial partner.
Visiting Researcher
Concordia University (Dec. 2024–Apr. 2025)
AIOps & Software Observability Research and Technology Lab, with Abdelwahab Hamou-Lhadj.
PhD, Software Engineering & Intelligent Systems
University of Alberta (2025)
Research in AIOps, supervised by James Miller and Abdelwahab Hamou-Lhadj.
Read my doctoral thesis ↗
Summer Research Intern
Polytechnique Montréal · SAET Research Lab (Apr.–Aug. 2022)
Worked with Mohammad Hamdaqa on AIOps for Digital Twin applications, supported by a competitive POINT Scholarship.
Service & Recognition
- IEEE Transactions on Software Engineering (TSE) — 2026
- ACM Transactions on Software Engineering and Methodology (TOSEM) — 2026
- ISSTA — Artifact Evaluation Committee, 2026
- AIware — Program Committee, 2026
- ICST — Program Committee, 2026
- ICSME, Visions and Emerging Results Track — Program Committee, 2026
- International Workshop on Agile Model-Driven Engineering at STAF — Program Committee, 2026
- PST — Organizing Committee Volunteer, University of Ottawa, 2026
- CASCON — Program Committee, 2025–2026
- WiSER (Women in Science, Engineering & Research) — Executive Committee Member & Event Lead, University of Alberta, 2024–2026
- ACM Transactions on the Web (TWEB) — 2025
- SPEC Research Group, DevOps Performance Working Group — Member, 2024–2025
Selected Awards
- Distinguished Artifact Reviewer Award (2026), ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA).
- Outstanding Reviewer Award (2026) — ACM Transactions on the Web, recognizing reviewing contributions in 2025.
- ACM Grant for ICPE (2024) — ACM/SPEC International Conference on Performance Engineering.
- POINT Scholarship (2022) — Polytechnique Montréal International Research Internship.
- Graduate Student Conference Award (2021) — Faculty of Graduate Studies and Research, University of Alberta.
- Best Student Presenter Award (2019) — Electrical and Computer Engineering, University of Alberta.
- Doctoral Recruitment Scholarship (2017–2018) — University of Alberta.
CONNECT